Home – Home – Jornals – Scientific journal “Vestnik NSUEM” 2026 number 3

Widely used traditional rule-based fraud monitoring systems, without ap-propriate modernization using artificial intelligence tools, are unable to effectively combat fraudulent activities, which increasingly rely on new technological advances. This article presents a systematic review of Russian literature devoted to the applicability of various machine learning models and algorithms in relation to fraud monitoring tasks. The purpose of the proposed study is to identify promising areas for modernizing traditional fraud monitoring systems based on the identification, systematization, ranking, and experimental testing of machine learning models and algorithms, as well as a comparison of the obtained results with similar studies by foreign authors, which were carried out through the analysis of the flow of foreign publications. The PRISMA protocol was used as a methodological foundation for creating the review. The initial sample included 853 publications from the Russian eLibrary and CyberLeninka databases. The final sample, after all stages of analysis using the PRISMA method, included 39 published papers. Sixteen models and algorithms were considered, including deep learning models (LSTM, GNN, CYCLEGAN), optimization algorithms (GA), hybrids of quantum computing with deep learning models, classical (Random Forest, Isolation Forest, Decision Tree) and statistical models (Clusterization, Regression, Bayes), gradient boosting (XGBoost, AdaBoost, CatBoost), and data balancing models (SMOTE, SMOTENN). An analysis of publications revealed that Random Forest (first place) and Logistic Regression (second place) are the most frequently used algorithms in building anti-fraud systems, while Isolation Forest and the Bayesian algorithm are among the least popular. The obtained results correlate with studies of the flow of English-language publications reflected in the works of foreign authors. As part of the experimental study, the first-place algorithm, Random Forest, was tested. Testing was conducted on an array of 2.2 million anonymized transactions of bank clients. As a result, the Recall metric reached 0.941, the F1-score was 0.9267, the ROC-AUC was 0.923, and the Precision was 0.9188. Accuracy was 0.9741. This is consistent with the results of similar experiments by foreign authors and provides practical confirmation of the preferable choice of the Random Forest algorithm for detecting fraudulent transactions.

The article is devoted to reducing the cognitive burden on human development in the field of spatial development of regions in the context of data-driven management. A model of a regional analytical center based on cooperative multiagency with LLM is proposed, including a formal description of the task, a project office of agents, a resource base (data, knowledge graphs, skills, rules) and a simulation model of the region. For transparency, Human in the Loop, LLM as a Judge, and Datalog validation schemes are used. A prototype has been developed for Java 25, Spring Boot 4, Angular 20, PostgreSQL 17. The scenario in the field of transportation in Buryatia shows that mutual verification of agents improves the quality of solutions developed for agricultural production.

This study investigates the computational efficiency of various optimization algorithms on the penalized QUBO (Quadratic Unconstrained Binary Optimization) formulation of the two echelon vehicle routing problem (2E VRP). The model includes variable and fixed transportation costs (c_fix = 50 000 RUB per trip for vehicle activation) and is tested on a forest industry logistics instance covering the Ural and Siberian Federal Districts (UFD+SFD) with 207 nodes, 19 constituent entities of the Russian Federation, and N = 10,162 binary variables. It is established that the genetic algorithm (GA), including variants with initialization from a known feasible solution, systematically fails to reach feasibility on penalized QUBO landscapes of this scale. The reason is structural: a zero diagonal in the dominant linearization block W (accounting for 87 % of the variables) and the trap of the trivial minimum x = 0. In contrast, simulated annealing (SA) on the same Q matrix finds feasible solutions in all runs, improving the objective function by 55 % relative to the constructive heuristic. A hybrid approach is proposed: a three stage constructive heuristic “Feasibility First” generates an initial feasible solution in 1.7 seconds, after which SA optimizes it in the vicinity of the feasible region. The resulting F_QUBO value is -13 197 240 (a gap of 11.6 % from the continuous relaxation) with all constraints fully satisfied. The reduction in total transportation cost (variable + fixed costs) compared to GA amounts to 65 %.

This article examines the coordination gap between monetary, fiscal, and investment policies in the contemporary Russian economy. It demonstrates that when the key rate, fiscal stimulus, and investment incentives are set in different directions, a mutual weakening of monetary and fiscal policies occurs. It also proposes an analytical framework, including a “macro-sustainability triangle” model, a coordination gap index, and a system of empirical equations for assessing the impact of monetary and fiscal stimulus on investment activity.

This article presents a comprehensive analysis of the structure and dynamics of the credit portfolio of the Russian banking sector in 2020-2024 and provides a forecast of its development up to 2030, taking into account key macroeconomic factors. The study is based on statistical and comparative analysis, as well as elements of economic modeling that reflect the impact of interest rates, inflation, and government support measures on different lending segments. A comparative analysis of credit portfolio structures in Russia, the United States, the Eurozone, and China reveals that structural differences are determined by economic models, the level of financial market development, and the degree of state involvement in resource allocation. The findings indicate that the credit portfolio structure reflects not only the current state of the economy but also its institutional characteristics. The proposed approach to credit portfolio forecasting, based on differentiated sensitivity of lending segments to macroeconomic factors, contributes to the development of methodologies for analyzing banking sector dynamics and can be applied in further research on financial stability and monetary policy.

The article is devoted to the role of non-financial reporting in ensuring transparency of enterprises’ activities. The author examines the features of standards for the preparation of non-financial statements, describes the existing problems in its formation. The article suggests the recommended structure of a non-financial report, which allows for the systematization of information disclosure, and analyzes the reports of Russian companies, revealing differences in the completeness of information disclosure across various industries and indicators. As a result, measures are proposed to improve standards, formats, and procedures for providing non-financial information.

The article presents a comparative analysis of statistical approaches to measuring credit activity: credit growth rates, the credit-to-GDP gap (with different smoothing parameters of the Hodrick-Prescott filter), and the credit impulse. The relevance of the study is driven by the need for timely identification of credit cycle phases for monetary and macroprudential policy purposes, especially amid structural shifts in the Russian economy. Based on data from the Bank of Russia and Rosstat for the period Q1 2010 - Q1 2025, an empirical calculation of the indicators is carried out, and their ability to identify periods of credit market overheating and post-crisis recovery is assessed. Recommendations are proposed for the application of these indicators in the monitoring practice of the Bank of Russia.

This article examines trends and developmental characteristics of key demographic processes in the BRICS member countries, particularly the total fertility rate. It also examines issues related to these countries’ contribution to global demographic processes, including population change, shifts in reproduction patterns, and other factors. Attention is paid to current trends shaping the global demographic situation.

This study investigates whether domestic investment or foreign direct investment plays a more decisive role in shaping Nigeria’s economic growth, while accounting for key macroeconomic conditions. The study employs an Autoregressive Distributed Lag (ARDL) model built on annual time series data for the period from 1980 to 2024. The results reveal that domestic investment exerts a positive and statistically significant effect on economic growth in both the short and long-run. Foreign direct investment stimulates growth in the short-run but shows a negative long-run association, suggesting weak sectoral linkages and limited structural transformation. Policy efforts should prioritize domestic resource mobilization, lower borrowing costs, and redirect foreign investment toward productivity enhancing sectors such as agriculture, manufacturing, information and communication technology, to ensure sustainable and inclusive economic growth in Nigeria.

The article examines the problem of amateurism in the field of statistics and its socio-economic consequences. It analyses the key manifestations of unprofessionalism, including the distortion of terminology, the incorrect use of data sources, the erroneous application of analytical methods, and the superficial interpretation of results. The study demonstrates how such practices lead to the proliferation of false beliefs, the adoption of inefficient management decisions, and the destabilisation of public consciousness. The paper substantiates the danger of an amateurish approach to statistics as a risk factor for the social and economic stability of society.

In a dynamically changing labor market with high volatility in job openings, standard job classifications that create intuitive job categories are often insufficiently detailed for in-depth economic and statistical analysis. A particular problem arises from labor market segments that are difficult to categorize using standard classifications, which, while accumulating a significant share of vacancies, defy simple classification, distorting the sectoral structure of labor demand. This article proposes and tests a method of contrasting multimodal graph clustering for detailing latent segments of regional labor markets. The methodology is based on a multi-stage approach involving normalization and vectorization of text data on job postings using the multilingual Sentence-BERT model, filtering semantic noise by calculating contrastive residual vectors relative to the centroids of known categories, enriching the feature space with multimodal data (salary level, work experience), constructing a weighted graph based on k-nearest neighbors and partitioning it into microclusters using the Leiden algorithm, followed by hierarchical aggregation of microclusters into macrosegments of a given size and balancing semantic proximity and weight constraints. The result of the work was the identification of interpretable latent segments within a heterogeneous category, often referred to as “other”: administrative and managerial personnel, blue-collar jobs, sales personnel, and highly specialized technical and production roles. The quality of the segmentation is confirmed by compliance with the specified weight constraints (target segment size of 5000 units) and the semantic interpretability of the resulting clusters based on c-TF-IDF. The proposed approach improves the analytical value of the data, automates the updating of the occupational classifier, and can be used to monitor structural shifts in the regional economy.

The relevance of developing formalized tools for assessing investment sustainability is driven by the increasing spatial differentiation of socio-economic development in the Russian Federation and the growing role of investment policy in ensuring long-term economic growth. Traditional rating methodologies used to compare the investment attractiveness of territories typically rely on the aggregation of a limited set of statistical indicators and do not allow for the identification of stable structural patterns in the formation of the investment potential of regional economies. The methodological foundation of the research is based on the application of multivariate statistical analysis methods, indicator normalization, and matrix procedures for aggregating regional data. The aim of the research is to formalize an integral index of investment sustainability as a function of a normalized vector of regional indicators and a system of weighting coefficients reflecting the relative significance of investment strategy factors. The results of the research: the constructed model accounts for key components of the regional investment environment, including investment potential, institutional conditions, infrastructure endowment, and the level of investment risks. To determine the structure of weighting coefficients, the use of principal component methods and entropy criteria for the informational significance of indicators is proposed, which ensures the statistical validity of the integral assessment. Practical significance: the presented model can consider the investment space of a regional economy as a multidimensional system of interrelated parameters, in which the sustainability of investment development is determined by the structure of statistical relationships between key socio-economic characteristics of the regions.